Abstract

A novel adaptive step length bacterial foraging algorithm (ASBF) for high-dimensional function optimization is presented in this paper. The proposed algorithm have some main advantages on traditional bacterial foraging algorithm such as adaptive step length and a new tumble method to overcome trapping  in local optima. The algorithm has been  evaluated on standard high-dimensional benchmark functions and compared with improved PSO and Genetic algorithms respectively. The simulation results have demonstrated fast convergence ability and improved optimization accuracy of ASBF.